Warning:
This page is
provided for historical and archival purposes
only. While the seminar dates are correct, we offer no
guarantee of informational accuracy or link
validity. Contact information for the speakers, hosts and
seminar committee are certainly out of date.
4:15pm, Adamson Wing, Baker Hall
This research was performed for my Ph.D until 1996 in Univ. of Tsukuba, Japan (Supervisor was Prof. Shin'ichi Yuta). In this talk, I introduce research approach "Task Orietned Approach" which we did, and reports one experimental result of autonomous navigation including door opening behavior.
Generally, robotic research is performed by following step, 1) To propose a general theory or algorithm for robot 2) To confirm an abailablity by simulation or experiment There are many researches along this approach, and many good theories and algorithms are proposed. However only basic behaviors are realized under strict condition, and intelligent versaile robot is a very long way, actually.
To grow out of this situation, we propose "Task Oriented Approach" for robotic research. First we set up a concrete task, and then we realize the task step by step. On the process of it, we may find new theory or algorithm which is directly connected with actual problems. We fixed our reserach task as "Autonomous Navigation including Door Open Behavior", in indoor environment.
We developed one robot system to realize the task, and realized robust navigation including door opening behavior. In this talk, I report both hardware and software issue of the robot, and experimental results. On the process of Experiment, we found not only know-how, but several knowledges and problems which common mobile manipulator has. We also discuss about them.
For appointments, please contact the speaker: keiji@ri.cmu.edu
This appears on the World Wide Web at http://www.frc.ri.cmu.edu/~mcm/seminar.html